PROJECT SUMMARY Much of late-life cognitive decline cannot be explained by Alzheimer?s disease (AD) or other common age- related neuropathologies. In fact, every individual is either resilient or susceptible to AD to a certain extent, due to their unique genetics and environment. Over the past 15 years our center has identified numerous environmental and psychological risk factors associated with faster or slower cognitive decline, and several molecular markers of resilience, that point to the existence of molecular networks that underlie resilience. The proposed project builds on this prior work. The overall goal of this proposal is to define the complex molecular basis of resilience to AD, using brains with various levels of resilience and in vitro human model systems, to identify novel therapeutic targets for cognitive decline. To do this, we will take a genome-wide approach to identify key molecular drivers of resilience in specific cell types in the human brain. Then we will perform rigorous tests of the molecules we have identified using brain cells from many different humans. Specifically we will see if we can genetically stimulate these cells to become more resilient to the effects of aging and Alzheimer?s disease. Two key innovations separate this project from previous work. The first aspect is our focus on individual cell types. Typically molecular measurements of brain data contain a mixture of dozens of cell types. We will measure each of these cell types individually, using single-cell RNAseq (scRNAseq), to identify which cell types are most related to resilience to AD. Within these specific cell types we will use computational network analysis to identify a smaller number of gene genes within the molecular systems affecting resilience. These predictions facilitate the 2nd aspect of this study which is unusual, which is our plan to change these genes in a human model of AD. The model we will use are neurons and glial cells derived from 50 individuals with different level of resilience to the common sporadic form of Alzheimer?s. We will use genome engineering to affect the abundance of genes that we predict are related to resilience in all of these cell lines. In this way we can check for resulting gene expression signatures of resilience as well as cellular phenotypes associated with health cognition, which persist in the face of AD pathology. The proposed project will deliver a comprehensive set of molecular networks and key molecules that underlie resilience to AD and other common brain pathologies. It will do so by breaking common barriers to progress in this area: 1) accurate identification of targets through a single cell approach and computational network methods, and 2) testing in realistic human models. The proposed project will provide high-confidence targets for therapeutic development. Thus, the proposal will have a strong and sustained impact on the field.